Skip to content

Latest commit

 

History

History
155 lines (108 loc) · 4.04 KB

README.md

File metadata and controls

155 lines (108 loc) · 4.04 KB

🛍️ ShopSmart Multi-Agent AI

ShopSmart Multi-Agent AI is an intelligent shopping assistant built with CrewAI, Google Gemini (Generative AI), and multi-agent orchestration.
It helps users in Egypt find the best products across Amazon.eg, Jumia, and Noon by:

  • 🔎 Searching e-commerce platforms
  • 📊 Comparing features and pricing
  • 💬 Analyzing customer reviews
  • 🏆 Delivering a final AI-driven recommendation

🚀 The project is fully deployed on Streamlit Cloud.

✨ Features

  • Multi-Agent Architecture powered by CrewAI:

    • 🔎 Search Specialist: Finds relevant product listings
    • 📊 Product Analyst: Compares features & pricing
    • 💬 Review Synthesizer: Summarizes customer feedback
    • 🏆 Recommendation Expert: Provides the final purchase recommendation

  • Real-Time Data Sources

    • Product listings from Amazon.eg, Jumia, and Noon
    • Web search integration via Tavily API
    • On-page scraping with CrewAI Tools

  • Generative AI (LLM)

    • Powered by Google Gemini (gemini-2.0-flash)
    • Used for reasoning, summarization, and recommendation generation

  • Deployed Application

    • Fully interactive Streamlit UI
    • Secure environment with .env and Streamlit Secrets

🖼️ Demo

🔍 Example Query

"Best budget smartphone with a good camera under 5000 EGP"

✅ AI Workflow

  1. Search → 2. Analyze → 3. Summarize Reviews → 4. Recommend

📊 Example Output

## 🏆 Final Recommendation  

**Product Name**: XYZ Budget Smartphone  

### ✅ Why We Recommend This  
- Affordable price with excellent camera performance  

### 👍 Pros  
- Great low-light photography  
- Long battery life  
- Affordable under 5000 EGP  

### 👎 Cons  
- Limited storage  
- Average display brightness  

### 🔗 Purchase Link  
[Buy Now](https://jumia.com.eg/xyz-smartphone)

🏗️ Project Structure

ShopSmart-MultiAgent-AI/
│── app.py                 # Streamlit app entrypoint
│── requirements.txt       # Dependencies
│── src/
│   ├── shopping_agents.py # Defines CrewAI agents
│   ├── shopping_tasks.py  # Defines CrewAI tasks
│── .env.example           # API keys template
│── README.md              # Project documentation

⚙️ Installation & Setup

1️⃣ Clone the Repository

git clone https://github.com/alaaashraf24/ShopSmart-MultiAgent-AI.git
cd ShopSmart-MultiAgent-AI

2️⃣ Create Virtual Environment

python -m venv venv
source venv/bin/activate   # Mac/Linux
venv\Scripts\activate      # Windows

3️⃣ Install Dependencies

pip install -r requirements.txt

4️⃣ Configure API Keys

Create a .env file (or use Streamlit Secrets):

GOOGLE_API_KEY=your_google_api_key
TAVILY_API_KEY=your_tavily_api_key

5️⃣ Run the App Locally

streamlit run app.py

🌐 Deployment

This project is deployed on Streamlit Cloud

🛠️ Tech Stack

📌 Key Learnings & Highlights

  • Designed a multi-agent AI pipeline with CrewAI (Agents + Tasks + Orchestration)
  • Integrated Google Gemini LLM for structured reasoning & natural language generation
  • Used Tavily API for web search and ScrapeWebsiteTool for extracting product data
  • Deployed a production-ready AI assistant on Streamlit Cloud
  • Applied prompt engineering and LLM chaining for reliable structured outputs

📄 License

This project is licensed under the MIT License.